JointResearchCentre

Advanced Non-animal Models in Biomedical Research

Breast Cancer

EUR 30334/1 EN

This publication is a JRC Technical report by the Joint Research Centre (JRC), the European Commission’s science and knowledge service. It aims to provide evidence-based scientific support to the European policymaking process. The scientific output expressed does not imply a policy position of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use that might be made of this publication. For information on the methodology and quality underlying the data used in this publication for which the source is neither Eurostat nor other Commission services, users should contact the referenced source. The designations employed and the presentation of material on the maps do not imply the expression of any opinion whatsoever on the part of the European Union concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries.

This report has been produced within the context of a scientific collaboration between the JRC (Directorate for Health, Consumers and Reference Materials, Chemicals Safety and Alternative Methods Unit) and the ALTER-N.A.M. consortium, who received financial support from the JRC (contract ref: JRC/IPR/2018/F.3/0035/OC).

This collaborative study was coordinated by Laura Gribaldo on behalf of the JRC’s EU Reference Laboratory for alternatives to animal testing (EURL ECVAM).

The collection of non-animal models described in this report is publicly available from the JRC Data Catalogue.

Contact informationEuropean Commission, Joint Research Centre (JRC), Chemical Safety and Alternative Methods Unit (F3)Via E. Fermi 2749, I-21027 Ispra (VA), ItalyJRC-F3-ENQUIRIES@ec.europa.eu

EU Science Hubhttps://ec.europa.eu/jrc

JRC122309

EUR 30334/1 EN

PDF ISBN 978-92-76-24689-3 ISSN 1831-9424 doi:10.2760/618741Print ISBN 978-92-76-24690-9 ISSN 1018-5593 doi:10.2760/180550

Luxembourg: Publications Office of the European Union, 2020

© European Union, 2020

The reuse policy of the European Commission is implemented by the Commission Decision 2011/833/EU of 12 December 2011 on the reuse of Commission documents (OJ L 330, 14.12.2011, p. 39). Except as otherwise noted, the reuse of this document is authorised under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/). This means that reuse is allowed provided appropriate credit is given and any changes are indicated. For any use or reproduction of photos or other material that is not owned by the EU, permission must be sought directly from the copyright holders.

All content © European Union, 2020, except: cover and p. ii and iii, ©Jo Panuwat D - stock.adobe.com; p. 3, graphic elaboration from ©Freepik - www.flaticon.com; p. 3, ©Africa Studio - stock.adobe.com; p. 7, ©MIMOHE - stock.adobe.com; p. 10, ©LIGHTFIELD STUDIOS - stock.adobe.com; p. 22, ©Africa Studio - stock.adobe.com; p. 23, ©Jo Panuwat D - stock.adobe.com; p. 25, ©Voyagerix - stock.adobe.com; p. 31, ©Gorodenkoff - stock.adobe.com.

Design and layout: Debora Valsesia and Adelaide Dura.

How to cite this report: Folgiero, V., Romania, P., Rossi, F., Caforio, M., Nic, M., Dibusz, K., Novotny, T., Busquet, F., Straccia, M. and Gribaldo, L., Advanced Non-animal Models in Biomedical Research: Breast Cancer, EUR 30334/1 EN, Publications Office of the European Union, Luxembourg, 2020, ISBN 978-92-76-24689-3, doi:10.2760/618741, JRC122309.

Advanced Non-animal Models in Biomedical Research

Breast cancer

Advanced Non-animal Models in Biomedical Research: Breast Cancerii

TABLEOF CONTENTS

Abstract 1

Acknowledgments 2

1 Introduction 4

1.1 Overview of breast cancer 4

1.2 Recent development in breast cancer research 5

2 Methodology 8

2.1 Selection criteria 8

2.2 Information sources 8

2.3 Systematic search 8

2.4 Method summary 9

Table of contents iii

3 Results and discussion 11

3.1 Publications employing human-based models 11

3.2 Fields of application of advanced models: from cancer pathogenesis to drug discovery 13

3.3 Distribution of models into categories 14

3.4 Increased use of 3D models 17

3.5 Relevance of advanced models for the disease features under analysis 18

4 Conclusions 23

5 References 26

6 Annex 32

Abstract 1

Abstract

The European Cancer Information System (ECIS) indicates that in the EU over 355,000 women were diagnosed with breast cancer in 2020 (13.3% of all cancer diagnoses).

Despite advances in early detection and better understanding of the molecular bases of breast cancer biology, approximately 30% of all patients with early-stage breast cancer have recurrent disease, which is metastatic in most cases.

To offer better treatment with increased efficacy and low toxicity, selecting therapies based on the patient and the clinical and molecular characteristics of the tumour is necessary.

Although preclinical breast cancer research relies heavily on animal models, experimental approaches that more accurately recapitulate breast cancer pathogenesis are still missing. Furthermore, in vivo models, mostly rodents, present some disadvantages: high costs; high-variability (due to lack of standardisation across laboratories); and most importantly the

physiological environment provided is not the human one.

Therefore, the JRC’s EU Reference Laboratory for alternatives to animal testing (EURL ECVAM) launched a study to survey the state of the art of human-based models for breast cancer described in the scientific literature from January 2014 to March 2019, by retrieving specific information from 935 selected peer-reviewed publications.

Advanced Non-animal Models in Biomedical Research: Breast Cancer2

Acknowledgments

The authors acknowledge the generous support provided by JRC colleague Adelaide Dura – the assistance she provided during the finalisation and editing of the report was greatly appreciated. Thanks also to Monica Piergiovanni for using her expertise in the field of organ-on-a-chip to extensively review the inventory.

We also would like to express our appreciation to Maurice Whelan, Head of Chemical Safety and Alternative Methods Unit, and Guy van den Eede, Director of Health, Consumer and Reference Material Directorate, for their valuable guidance and input.

Advanced Non-animal Models in Biomedical Research: Breast Cancer4

1 Introduction

Breast cancer is the most commonly diagnosed cancer among women worldwide. Although there have been several breakthroughs in the treatment of breast cancer in the past few decades, the high incidence of relapse and progression after conventional therapies is deeply concerning and indicates a great need for developing new therapeutics (Ju et al., 2018).

1.1 Overview of breast cancer

Breast cancer is curable in about 70–80% of patients with early-stage, non-metastatic disease. Advanced breast cancer with distant organ metastases is considered incurable with currently available therapies.

On the molecular level, breast cancer is a heterogeneous disease; molecular features include activation of human epidermal growth factor receptor 2 (HER2, encoded by ERBB2), activation of hormone receptors (oestrogen receptor and progesterone receptor) and/or BRCA mutations. Treatment strategies differ according to molecular subtype.

The impressive increase in knowledge in the field of molecular biology and immunology

has helped to elucidate the molecular characteristics of cancer representing the basis for a plethora of upcoming drugs. However, although important improvements have been achieved in recent years in terms of metastatic breast cancer outcomes, more and better treatments are needed. Additionally, the mechanisms underlying tumour resistance and how to overcome it are main topics of ongoing research.

Knowing the driving pathway at every given moment will enable the correct determination of the optimal sequence of therapies, which currently is largely unknown for all advanced breast cancer subtypes.

All existing therapies hit less than 500 molecular targets (Drews, 2000) suggesting that there are many unexplored targets for drug discovery within the human interactome that comprises possibly 1 million proteins and over 1 trillion potential interconnections. Nonetheless, there are clearly other limitations in drug development. Less than 10% of investigational drugs based on new molecules proceed beyond early development (Kola and Landis, 2004); the approval rate for new oncology drugs is about 5% (Kinders et al., 2007).

Perhaps the lack of significant progress partly reflects the drug development process in which preclinical animal models play a central role. The leading causes of attrition of new drugs are generally cited as being unpredictable toxicities and lack of efficacy, the early identification of which are primary goals for preclinical animal models.

Furthermore, as breast cancer is a highly heterogeneous disease, heterogeneity is often evident even within the same tumour. Cell line xenografts and genetically manipulated mouse models are more homogeneous. The relative homogeneity of these models may

Less than 10% of investigational drugs based on new molecules proceed beyond early development

Introduction 5

render them overpredictive or underpredictive, depending on how prevalent their phenotype is in the human disease. A prediction of high sensitivity often leads to a drug being considered a strong target for human testing.

Overpredicted drugs would show limited activity/potency in human efficacy trials and so (after significant investment) experience a high attrition rate if the animal models were too sensitive. While current breast cancer models may well overpredict sensitivity relative to the human disease, it is difficult to assess underprediction because the lack of activity in preclinical animal models could lead them to be dropped early.

1.2 Recent development in breast cancer research

In recent years, great progress has been made in the molecular target therapy of breast cancer, representing the pioneering field of precision medicine. Trastuzumab is regarded as the cornerstone of targeted therapy in HER-2-positive breast cancer and shows considerable efficacy in both neoadjuvant and adjuvant therapy (Ross et al., 2009; Ishii et al., 2019; Zimmer and Denduluri, 2019).

CDK4/6 and mTOR inhibitors exhibit the ability to reverse the resistance to endocrine and targeted agents to some extent (Yardley et al., 2013; Goel et al., 2017). PARP inhibitors also show immense potential in the treatment of the BRCA1/2-mutated subgroup (McCabe et al., 2006). Given that the hallmark of breast cancer is the great heterogeneity in which biomarkers are diverse not only between primary and metastatic tumours, but also within a single tumour or during tumour progression, in recent years surrogate intrinsic tumour phenotypes have also been used for treatment individualisation.

At research level, efforts continue for a better understanding of the biological heterogeneity

of breast cancer, as well as of mechanisms of tumour resistance and biomarkers predictive of response to the different therapeutic options. The generation of primary cell lines from human tumour biopsy is highly informative to identify novel biomarkers for the development of personalised “antigen-specific antibody”.

This approach overcomes the use of animal models for xenograft injection of commercial tumour cell lines, which partially resemble the real expression of breast cancer biomarkers. In addition, xenografts using human cell lines to test drug responses do not often correlate with clinical activity in patients. This is due to immunological deficits such as loss of T- and B-cell responses, depending on the selected mouse models (Richmond and Su, 2008). Moreover, the selection of cancer stem cells for in vitro cell culture from tumour biopsy and their protein and genetic characterisation is an exhaustive method to characterise each tumour population with the aim to develop specific therapy (Jin et al., 2017).

Recent studies reported the role of inflammation in the development of breast cancer, supporting the evidence that tumorigenic signalling pathways are not sufficient for explaining a complete breast tumour progression. In this

The generation of primary cell lines from human tumour biopsy is highly informative to identify novel biomarkers for the development of personalised “antigen-specific antibody”

Advanced Non-animal Models in Biomedical Research: Breast Cancer6

contest also the role of the immune system is crucial. During the onset of breast cancer, immune system cells like lymphocytes, NK cells and macrophages produce inflammatory factors in the tumour microenvironment. Some of these factors are cytokines, like TNF-a, TGF-b, and interleukins, exerting a pivotal role in tumour progression and contributing to the induction of proliferation, angiogenesis and epithelial-mesenchymal transition (Bahiraee et al., 2019).

Breast cancer research is mostly carried out using animal models, which has been instrumentally useful so far since they can provide a physiologically relevant microenvironment and an intact immune system. From this perspective, many animal models have been generated. We can classify the in vivo models either as Genetic ally Engineered Mouse Models (GEMM, either conventional or conditionals), as Xenograft based (Cell-line Derived Xenograft, CDX; and Patient-Derived Xenografts, PDX) and as Syngeneic-transplant based (Park et al., 2018).

Although preclinical breast cancer research has relied until now on in vivo models, approaches that accurately recapitulate breast cancer pathogenesis are still missing (Manning et al., 2016). Furthermore, in vivo models also present certain disadvantages. In fact, in vivo murine models have: high costs; high-variability, due to lack of in vivo models standardisation across laboratories; and most importantly the physiological environment provided is the mouse microenvironment (Holen et al., 2017), not the human.

In order to explore the trends of human in vitro and in silico-based models in breast cancer research, we present here the results of the systematic literature review of 935 scientific peer-reviewed articles, published from January 2014 to March 2019, using non-animal methods in breast cancer research, retrieved in PubMed, Scopus and Web of Science databases.

Advanced Non-animal Models in Biomedical Research: Breast Cancer8

2 Methodology

The review strategy employed retrieved 119,722 candidate abstracts. After a selection based on titles and abstracts, 48,327 scientific articles were retrieved for the full-text selection.

The full-text analysis resulted in a selection of 935 articles, from which all the identified data were extracted and analysed.

2.1 Selection criteria

The systematic search strategy considered any scientific article describing or dealing with in vitro human models or methods or assays or test systems in the field of breast cancer research, based on the dynamic classification shown in Annex - Table 1, as inclusion criteria.

In addition, it was considered as inclusion criteria any scientific article describing or dealing with any in silico model, such as an algorithm or mathematical or computational simulations.

The following initial set of flagged search terms was determined as inclusion search terms, for the publications retrieval based on title/abstract analysis:

model* OR assay* OR “test* system*” OR “in vitro” OR “ex vivo” OR in-vitro OR ex-vivo OR organoid* OR spheroid* OR 3D OR coculture OR co-culture OR microfluidic* OR microphys* OR biops* OR explant* OR “cell culture” OR “stem cell*” OR stem-cell* OR “primary culture” OR simulation* OR algorithm* OR mathematic* OR computation* OR chip

The search strategy proposed considered the exclusion criteria listed in Annex - Table 2 and the following initial set of flagged search terms were determined as exclusion search terms for the publications retrieval based on title/abstract analysis:

“mouse model” OR murine OR mice OR rat OR rats OR “Controlled Study” OR “Priority Journal” OR “Major Clinical Study” OR “Animal Experiment” OR “Animal Model” OR “Animal Tissue” OR “Prognosis” OR “Follow Up” OR “Follow-Up” OR “Retrospective Stud*” OR “Prospective Study” OR “Case Control Study” OR “case stud*” OR “case-stud*” OR “Nude Mouse” OR “Psychology” OR review OR “Case Report” OR questionnaire* OR “Diagnostic Imaging” OR “Mammography” OR cross-sectional OR survey* OR “Meta-Analysis” OR “meta-analysis” OR hiv OR infection* OR aids OR hepatitis OR influenza OR “clinical trial*” OR xenotransplant* OR xenograft* OR papilloma* OR gvhd OR “qualitative study” OR workshop OR sympos* OR “conference* proceeding*” OR cohort OR descent OR ancestr* OR participant* OR population OR gwas OR “genome wide analysis” OR “methyl* analys*” OR polymorphism*

2.2 Information sources

To perform the systematic literature search, it was agreed to focus on human-based models published in the last five years (January 2014 up to March 2019). In order to generate the most inclusive datasets, multidisciplinary citation databases and indexing services (Web of Science and Scopus) and the specific biomedical sciences citation database, PubMed, were used. Furthermore, grey literature sources of information were monitored to retrieve news and/or highlights on non-animal methods in the field (Annex - Table 3).

2.3 Systematic search

We finally retrieved 935 full-texts from where the data were extracted and analysed. However, to end with the selected full-texts,

Methodology 9

we applied five sequential strategies (Annex - Table 4) as illustrated in Figure 1.

We initially retrieved a total number of 119,722 scientific peer-reviewed journal articles by applying Strategy A. After the selection based on titles and abstracts applying Strategies B and C, we finally sorted out 48,327 publications for full text review. During the analysis of the 48,327 publications, we observed a significant risk of redundancy for few models, especially for immortalised cell lines. We managed to partially override such redundancy by designing a new strategy applied to abstracts (strategy D).

However, such high-represented models are still over represented in the repository due

1 https://europa.eu/!bM83pv

to lack of their specific description in the abstract texts. We therefore designed a new strategy (strategy E) to specifically retrieve peer-reviewed publications reporting new models, lowering redundancy of high literature represented models.

2.4 Method summary

The data from the scientific articles were extracted based on the method-summary format including fields that are reported in Annex - Table 5. The resulting collection of advanced non-animal models is publicly available from the EURL ECVAM collection in the JRC Data Catalogue1.

Figure 1: Selection process.

Results and discussion 11

3 Results and discussion

3.1 Publications employing human-based models

After retrieving a total of 119,722 abstracts following our study criteria (see Section 1), we have selected and analysed 935 peer-reviewed articles published from January 2014 to March 2019. The vast majority of articles (924) dealt specifically with breast cancer and only 11 articles were focused on the role of inflammation in breast cancer research (Figure 2). The number of journal articles published increased each year and,

in 2018, 214 publications reported the use of non-animal models compared to the 129 articles of 2014 (Figure 2) which is the 66% more within 4 years.

The majority of peer-reviewed articles focused on breast cancer initiation and development (384 articles; Figure 3), including paramount processes like epithelial–mesenchymal transition, cancer stem cell (CSC) formation, metastatic behaviour or mutation functional significance (Mylona et al., 2014; Feng et al., 2015; Avivar-Valderas et al., 2018; Bocci et

Figure 2: Distribution of peer-reviewed articles by year of publication from January 2014 to March 2019. Articles are classified for their main focus: BC, Breast cancer; IBC, inflammatory breast cancer; or both IBC/BC.

Advanced Non-animal Models in Biomedical Research: Breast Cancer12

al., 2019). A total of 210 scientific articles employed non-animal models to develop and test pharmacological and/or physical treatment to stop or revert breast cancer pathogenesis (Figure 3), such as small molecules characterisation and testing and large-scale high-content screening based on advanced culture systems (Härmä et al., 2014; Li et al., 2016). Moreover, microenvironment-tumour interactions were studied in 75 publications (Figure 3), also using lab-on-chip technologies (Choi et al., 2015; 2016).

Forty-nine articles used non-animal models to study metastasis, reporting the development of new cell substrates able to prompt invasion and migration mimicking in vivo conditions (Cavo et al., 2018). A total of 36 articles focused on studying metastasis microenvironment, also

presenting new models potentially scalable to high-throughput (Nagaraju et al., 2018) or describing specific players in the tumour-stroma interaction (Hohensee et al., 2017). Non-animal models were also employed in 28 publications to dissect breast cancer initiation and development with the aim of testing pharmacological and/or physical treatment (Figure 3), e.g. to evaluate photodynamic therapy (Yang et al., 2015), or chemical drug efficiency to inhibit stem/progenitor proliferation (Farnie et al., 2014).

Furthermore human-based models were used in 30 publications for tumour detection and classification for clinical stratification purposes, including emerging methods such as circulating tumour cells analysis (Hainsworth et al., 2016) or in silico prediction models

Figure 3: Number of peer-reviewed publications focused on one or more breast cancer disease features. E.g.: 384 publications out of 430 were focused on breast cancer initiation and development only, the remaining 46 publications were focused on breast cancer initiation and development plus other disease features.

Results and discussion 13

(Karapanagiotis et al., 2018). The remaining studies were tackling many other aspects of breast cancer pathogenesis and therapeutic strategies (Figure 3).

3.2 Fields of application of advanced models: from cancer pathogenesis to drug discovery

In our analysis we identified eight areas of application of non-animal models in breast cancer research. They are listed in Figure 4.

More than half of the publications focused on certain aspects of breast cancer pathogenesis to identify or to model specific disease mechanisms (51.2%; Figure 4A). This area showed an annual increase in the number of publications, which reached 108 articles in 2018 (Figure 4B). A total of 29.5% of journal articles reported the use of advanced models in studies on drug development and testing of

their efficacy against breast cancer (Figure 4A), using chemical libraries and not-directed to a specific target (Chen et al., 2018) or directed to specific cellular players (Patidar et al., 2016).

During the 5-year period under analysis, the interest in using non-animal models for drug development in breast cancer increased overtime, almost doubling the number of publications (36 articles in 2014 vs. 70 in 2018; Figure 4B). Publications reporting the development in breast cancer research of new models, such as new cell lines (Ali et al., 2017), or new techniques, represented 10.9% of all the retrieved articles (Figure 4A), with 14 articles published in 2018 and a median of 21 articles from 2015 to 2018 (Figure 4B).

A remaining 8.3% of publications dealt with the theoretical development of non-animal models including models aiming to predict cell survival to therapy (2.7%), both physical (Sung et al., 2018) and chemical (Lucantoni et al., 2018), breast cancer diagnostic applications

Figure 4: Eight applications for non-animal models in breast cancer research were identified. Panel A shows the percentage of each reported application in all retrieved articles, including publications from 2019. Panel B shows the distribution of articles by model application in breast cancer research from January 2014 to December 2018. The number of articles per year for the three major applications are shown.

Advanced Non-animal Models in Biomedical Research: Breast Cancer14

(2.0%), development of disease therapies (2.0%), qualification for specific applications (1.3%) (Figure 4A). Finally, 3 articles (0.3%) described their development both theoretically and experimentally.

3.3 Distribution of models into categories

The analysis of the scientific literature found that on the one hand 91% out of 935 selected publications used in vitro models for breast cancer research (Figure 5A), with a clear trend increasing overtime (Figure 5B). On the other hand, only a 5% were in silico models (Figure 5A). Thirty-eight articles (4%) dealt with models integrating in silico and in vitro (Figure

5A). The publication of articles presenting such in vitro/in silico models increased from the average of 5 articles per year to 15 articles in 2018 (Figure 5B). Whereas the use of in silico models alone were reported in a maximum of 12 publications in 2017 to a minimum of 6 articles in 2018 (Figure 5B).

The in silico models reported in 49 articles were mainly computational models and algorithms for cancer initiation and development studies and for tumour detection and classification (Figure 5C) (Greenbaum et al., 2014; Chapa et al., 2016). Thirty-eight publications using both in vitro and in silico methods were mainly integrating breast cancer cell lines-based models in combination with an in silico model (31 articles, Figure 5D). These

Figure 5: Panel A shows the percentage for each model category in all retrieved 935 articles. Panel B shows the distribution of peer-reviewed articles by year of publication from January 2014 to March 2019. Panel C shows the number of peer-reviewed publication for each type of in silico model/method. Panel D shows the number of peer-reviewed publications reporting the use of both in vitro and in silico human-based models.

Results and discussion 15

publications were mainly focused on breast cancer initiation and development (17 articles) and also on pharmacological and physical treatment of breast cancer cells (14 articles). Those studies that combine machine learning and in vitro approaches to identify new marker for evaluation in risk assessment (Ren et al., 2018) or those developing mathematical model to study tumour evolution (Reiter et al., 2018) are of particular interest.

Focusing on breast cancer research, in vitro studies from 2014 to 2018 used mainly cell-based models with a constant increase year after year (Figure 6A), representing 84% of total retrieved articles (Figure 6B). The use of human ex vivo models was reported in

107 peer-reviewed scientific articles during the period under analysis (Figure 6A), which corresponds to 12% of the total in vitro models we analysed (Figure 6B). Most ex vivo models were patient biopsies (88 articles), mainly solid biopsies used for the identification of biomarkers (Vici et al., 2014) or clinical applications, e.g. immunotherapy (Chen et al., 2019), while liquid biopsies were less represented (Bingham et al., 2017; Zhong et al., 2018) (Figure 6C).

Organ slice, intended as explanted tumour cut to be used for in vitro studies (Carranza-Torres et al., 2015), were reported in 15 publications (Figure 6C).

Figure 6: From January 2014 to March 2019, a total of 891 publications were reporting the using of in vitro models in breast cancer research. Panel A shows the distribution of peer-reviewed articles by year of publication from January 2014 to March 2019 and by type of in vitro model; values for the three most reported categories are shown per each year. Panel B shows the percentage for each category with respect to total in vitro models. Panel C shows the number of peer-reviewed articles per each type of ex vivo model.

Advanced Non-animal Models in Biomedical Research: Breast Cancer16

The use of both cellular and ex vivo models in the same study represents another research approach (Chi et al., 2017), with 2% of publications reporting this strategy to address their research hypothesis (Figure 6A and B). Cell free approaches alone or in combination with cells were used in few cases (2.1% in total), representing a small niche of in vitro models, but including a large spectrum of disease areas spanning from the mutational landscape (Kirkizlar et al., 2015) to the development of immunosensors (Eletxigerra et al., 2016) (Figure 6A and B).

A total of 801 out of 935 publications dealt with cellular in vitro models. Due the importance of this category, we disaggregated the data to provide a better view of cellular models in use. In 82% of the publications using human cell-based models, immortalised breast cancer cell lines were employed (Figure 7A). A total of 10% of the publications reported the use of immortalised cell lines and stem cell-like in vitro models and a 5% of research articles reported the use of human primary cultures (Figure 7A). Twenty-two articles (3%) reported the use in combination of immortalised cells and primary cell culture, while 8 articles (1%)

Figure 7: From January 2014 to March 2019, a total of 801 publications reported the use of immortalised cell lines in breast cancer research. Panel A shows the percentage for each type of human cell-based models category with respect to total number of cellular models. Panel B shows the distribution of peer-reviewed articles dealing with different type of human cell based-models by year of publication from January 2014 to March 2019; values for the three most reported cell-based models category are shown per each year. Each bubble, in panel C, represents a human immortalised cell line. The area of each bubble is proportional to the number of publications employing the specific immortalised cell line, the 12 most abundant human immortalised cell lines are indicated with the corresponding absolute number of references.

Results and discussion 17

used as experimental paradigms multiple cellular models (Figure 7A).

Immortalised cell lines reporting in journal articles increased from 87 publications in 2014 to 156 in 2018 (Figure 7B). Whereas, the number of publications using immortalised cells and stem cell-like models increased, during the same period, more than three times from 7 to 23 research articles (Figure 7B). On the other hand, research articles reporting primary cell culture showed relatively constant rate of publications in our 5-year analysis (Figure 7B).

Regarding breast cancer immortalised cells, we observed hundreds of different cell lines employed in several experimental paradigms (Figure 7C). Breast cancer cells lines included cells derived by different sources and cells genetically modified from original cell lines with similar genomic background. The most common cell lines reported in peer-reviewed

publications from January 2014 to March 2019 were in order of frequency of use: MCF-7; MDA-MB-231; T-47D; SKBR3; MCF 10A; BT-474; MDA-MB-468; BT-549; ZR-75-1; Hs 578T; MDA-MB453; SUM149; SUM159 (Figure 7C).

3.4 Increased use of 3D models

Culture conditions are very important to understand whether disease modelling is able to mimic as much as possible the physiological microenvironment of breast cancer. A better modelling of physiological conditions enhances the reliability of in vitro results.

A total of 83.2% (655 articles) of all publications employing cell-based models, reported cell cultures of individual immortalised cell populations and stem cell-like and primary cells (Figure 8). Only 9.6% (77 articles) of the published studies employed co-

Figure 8: Type of human cell-based model by culturing conditions.

Advanced Non-animal Models in Biomedical Research: Breast Cancer18

culture systems, e.g. to study chemotaxis of metastatic cells or to model the crosstalk with tumour microenvironment (Chung et al., 2017; Daubriac et al., 2018), and 3% (24 articles) of articles reported the use of microphysiological systems (MPS), mostly microfluidics culture models for 3D (Sabhachandani et al., 2016) and 2D (Ren et al., 2018) applications (Figure 8). The remaining 4.2% of studies (33 articles) employed several cell-based systems and culture conditions.

Less than half of the individual cell populations were cultivated in 2D (347 articles) and 33.2% were cultivated in 3D conditions (266 articles; Figure 8), where the use of scaffolds (63 articles) was the main approach, e.g. for the study of matrix topography (Riching et al., 2014; Clay et al., 2016) or microenvironment and metastasis (Eslami Amirabadi et al., 2017), followed by organoids (15 articles), including patient-derived organoids (Shirure et al., 2018) and spheroids (15 articles), also used in microfluidic systems (Xia et al., 2017).

In 7.4% of studies it is reported the use of both 2D and 3D models (59 articles; Figure 8), eventually comparing the two models or to highlight the value of their complementary use (Breslin and O’Driscoll, 2016; Yildiz-Ozturk et al., 2017). Moreover 9.7% of publications reported extracellular matrix-embedded conditions (2.5D) for individual cell cultures (78 articles; Figure 8). On the other hand, cellular co-culture models were almost equally used in 2D and 3D culture systems, while MPS were mainly used in 3D conditions and a minor fraction in 2D (Figure 8). A total of 39 articles (4.9%) reported the use of two individual cell populations cultivated both in 2.5D and 3D conditions (Figure 8).

The analysis of data collected showed that 3D models increased over time and, in 2018, they equaled the number of publications reporting

2 Throughput is defined as the number of samples that can be processed in parallel.

3 Content is defined as the quantity of information retrieved by each sample with a single analysis or method.

2D model implementation (Figure 9A). A small increase in the use of multiple models in the same research article has been observed, from 2014 to 2018, especially in 3D conditions (Figure 9A).

Spheroids, formed by single cell population or in co-culture with other cells, were the most employed 3D models (278 articles), followed by the use of scaffolds (Figure 9B). Of note, the use of spheroids and scaffolds in microfluidic systems was also reported in 41 peer-reviewed publications.

3.5 Relevance of advanced models for the disease features under analysis

Of all retrieved publications reporting non-animal models 22.4% focused on pharmacological and physical treatments of breast cancer and 41% studied the breast cancer initiation and development mechanisms (Figure 3).

To analyse these features, the throughput2 (productivity or automatisation) and the content analysis are key to screen several treatments in many different models and to retrieve as much molecular signalling information as possible. Hence, we screened the articles to determine the level of throughput and content of analysis3 and we found that 770 articles described low-throughput use of models, 112 articles a medium-throughput level and 50 articles a high-throughput usage (Figure 10).

From a content perspective, 835 studies applied low content analysis methods, whereas 83 were reporting high-content analysis (Figure 10). This resulted in 73.6% of articles reporting both low-throughput use and low-content analysis (Figure 10). On the one hand, 4.1% of the publications reported high-

Results and discussion 19

Figure 9: Distribution of articles by year of publication and by culture dimension. A total of 188 articles reported the use of multiple models with different culture dimensions. Each dot represents a peer-reviewed publication.

throughput use of non-animal models, but low content analysis. On the other hand, 7.3% of publications employed high content methods of analysis, but low throughput (Figure 10). Only 1.1% of the publications employed human-based models in high-throughput and applying high-content methods of analysis (Figure 10).

To understand whether non-animal models were relevant to study a specific disease feature, we analysed each article taking into consideration the type of human-based model used to address a disease feature (Figure 11). Seventy-nine percent of all retrieved publications implemented cellular models, where 36.6% were used to study breast cancer initiation and development and 21% were employed to test pharmacological and physical treatments (Figure 11). Cell-based models were also the main models to study metastatic processes and microenvironment-

tumour interactions, representing 7.8% and 6.8% of all studies, respectively (Figure 11). In the 4.3% of all peer-reviewed publications, ex vivo models were employed to study breast cancer initiation and development (Figure 11).

Then, we analysed the relevance of each human-based model with respect to the target disease feature under consideration. We found 630 peer-reviewed publications where the models had direct relevance for the disease feature that was studied (Figure 12). On the one hand, direct relevance was observed in 28.3% of the articles studying breast cancer initiation and development, 17.3% of the articles dealt with pharmacological or physical treatment, 6,4% focused on metastasis and 6,4% focused on microenvironment-tumour interactions. 2.7% of total publications were having a direct relevance for their use in tumour detection and classification (Figure 12). On the other hand, we determined that 258 scientific articles

Advanced Non-animal Models in Biomedical Research: Breast Cancer20

Figure 10: Number of articles by their throughput and analysis of content level.

Figure 11: Number of articles and their distribution by the type of model used to address specific breast cancer features. Only percentage greater than 1% of the total retrieved peer-reviewed publications are shown.

Results and discussion 21

employed non-animal models in a supportive manner for the specific disease feature they addressed (Figure 12). This was assessed in 14.4% of the total retrieved publications for studying breast cancer initiation and development, 6.8% of the articles to test pharmacological or physical treatments, while

4% of the publications dealt with metastasis and microenvironment-tumour interactions (2.4% and 1.6% respectively) (Figure 12). We were not able to determine the relevance of 47 publications, correspondong to 5% of the total publications, mostly dealing with cancer initiation and development (Figure 12).

Figure 12: Number of articles and their distribution by the relevance for the use of non-animal modelNAMs in studying specific breast cancer features.

Conclusions 23

4 Conclusions

The European Cancer Information System (ECIS) indicates that in the EU over 355,000 women were diagnosed with breast cancer in 2020 (13.3% of all cancer diagnoses). Although 5% to 10% of breast cancer are likely hereditary, this high incidence highlights the lack of knowledge about the events triggering breast cancer initiation and whether it skips immunological recognition.

This systematic review of human-based models for breast cancer resulted in 935 articles using in vitro and/or in silico models to study 18 different aspects of breast cancer pathogenesis. The main disease features addressed were breast cancer initiation and development at cellular levels by using mainly immortalised cell lines in particular, and ex vivo models such as tumour biopsies. The main interest in breast cancer initiation and development among the research community mirror the great importance in discovering the molecular bases of the starting events.

The incidence of female breast cancer increased steadily from 1970 to 2016 (WHO 2020) and the biomedical research also has increased its efforts. Our 5 year-time analysis captured a similar trend with peer-reviewed publications, with the use of non-animal models increasing each year. In parallel we also observed an increase in their usage for drug development and testing, which indicates the importance of finding new and more effective treatments for breast cancer. On the other hand, the publication of new human-based models for breast cancer

research was steady. However, the use of 3D models has increased suggesting a better approximation to breast cancer physiology, especially through the use of spheroids.

Our analysis showed that the implementation of human-based in silico models in breast cancer research is still a minor niche, but applied to all disease features. It is also worth pointing out the small group of publications reporting the use of both in vitro and in silico models, which has great potential.

Human breast cancer immortalised cells end up being the undisputable most used model reported in the systematic review. In fact, hundreds of cell lines have been found as reported in 747 peer-reviewed publications, including genetic engineered lines derived from commercially available or already qualified cell lines. They have been employed to study breast cancer initiation, treatments, metastatic process and the microenvironment-tumour interactions using several culturing conditions.

Regarding the selection of immortalised cell lines for modelling metastatic breast cancer an insightful evaluation has been made by an integrative analysis of genomic data (Liu et al., 2019). However, as in previous studies, most publications reported a low-throughput and low-content analysis. This was interesting since most of the models were immortalised cell lines in relative standard culture conditions, which would have allowed process automation and the analysis through omics.

Advanced Non-animal Models in Biomedical Research: Breast Cancer24

Considering the overall results of the review, the main conclusions are the following:

The use of non-animal models in human breast cancer research is extensive, especially in relation to cell-based in vitro

models.

The human-based models are mostly applied to elucidate disease mechanisms and to test drug candidates. In particular,

human-based models are focused on studying breast cancer initiation and development and on pharmacological or physical treatments.

Qualified and commercially available human breast cancer immortalised cell lines represent the most common models.

In vitro immortalised cells are often cultured in 3D conditions, such as spheroids, to better mimic breast cancer

pathophysiology.

Two-thirds of publications employed non-animal model as a relevant model to study the disease features of interest in breast

cancer research.

Advanced Non-animal Models in Biomedical Research: Breast Cancer26

5 ReferencesAli, R., Samman, N., Al Zahrani, H., Nehdi, A., Rahman, S., Khan, A. L., Al Balwi, M., Alriyees, L. A., Alzaid, M.,

Al Askar, A. and Boudjelal, M. (2017). Isolation and characterization of a new naturally immortalized human breast carcinoma cell line, KAIMRC1. BMC Cancer, 17, 803, doi:10.1186/s12885-017-3812-5.

Avivar-Valderas, A., McEwen, R., Taheri-Ghahfarokhi, A., Carnevalli, L. S., Hardaker, E. L., Maresca, M., Hudson, K, Harrington, E. A. and Cruzalegui, F. (2018). Functional significance of co-occurring mutations in PIK3CA and MAP3K1 in breast cancer. Oncotarget, 9, 21444–21458, doi:10.18632/oncotarget.25118.

Bahiraee, A., Ebrahimi, R., Halabian, R., Aghabozorgi, A. S. and Amani, J. (2019). The role of inflammation and its related microRNAs in breast cancer: A narrative review. J. Cell. Physiol., 234, 19480–19493, doi:10.1002/jcp.28742.

Bingham, C., Fernandez, S. V., Fittipaldi, P., Dempsey, P. W., Ruth, K. J., Cristofanilli, M. and Alpaugh, R. K. (2017). Mutational studies on single circulating tumor cells isolated from the blood of inflammatory breast cancer patients. Breast Cancer Res. Treat., 163, 219–230, doi:10.1007/s10549-017-4176-x.

Bocci, F., Gearhart-Serna, L., Boareto, M., Ribeiro, M., Ben-Jacob, E., Devi, G. R., Levine, H., Onunchic J. N. and Jolly M. K. (2019). Toward understanding cancer stem cell heterogeneity in the tumor microenvironment. PNAS, 116, 148–157, doi:10.1073/pnas.1815345116.

Breslin, S. and O’Driscoll, L. (2016). The relevance of using 3D cell cultures, in addition to 2D monolayer cultures, when evaluating breast cancer drug sensitivity and resistance. Oncotarget, 7, 45745–45756, doi:10.18632/oncotarget.9935.

Carranza-Torres, I. E., Guzmán-Delgado, N. E., Coronado-Martínez, C., Bañuelos-García, J. I., Viveros-Valdez, E., Morán-Martínez, J. and Carranza-Rosales, P. (2015). Organotypic Culture of Breast Tumor Explants as a Multicellular System for the Screening of Natural Compounds with Antineoplastic Potential. Biomed Res. Int., 2015, 1–13, doi:10.1155/2015/618021.

Cavo, M., Caria, M., Pulsoni, I., Beltrame, F., Fato, M. and Scaglione, S. (2018). A new cell-laden 3D Alginate-Matrigel hydrogel resembles human breast cancer cell malignant morphology, spread and invasion capability observed “in vivo.” Sci. Rep., 8, doi:10.1038/s41598-018-23250-4.

Chapa, J., An, G. and Kulkarni, S. A. (2016). Examining the Relationship between Pre-Malignant Breast Lesions, Carcinogenesis and Tumor Evolution in the Mammary Epithelium Using an Agent-Based Model. PLoS One, 11, e0152298, doi:10.1371/journal.pone.0152298.

Chen, I. X., Chauhan, V. P., Posada, J., Ng, M. R., Wu, M. W., Adstamongkonkul, P., Huang, P., Lindeman, N., Langer, R. and Jain R. K. (2019). Blocking CXCR4 alleviates desmoplasia, increases T-lymphocyte infiltration, and improves immunotherapy in metastatic breast cancer. PNAS, 116, 4558–4566, doi:10.1073/pnas.1815515116.

Chen, L., Long, C., Youn, J. and Lee, J. (2018). A Phenotypic Cell-Binding Screen Identifies a Novel Compound Targeting Triple-Negative Breast Cancer. ACS Comb. Sci., 20, 330–334, doi:10.1021/acscombsci.8b00026.

Chen, Y.-C., Zhang, Z., Fouladdel, S., Deol, Y., Ingram, P. N., McDermott, S. P., Azizi, E., Wicha, M. S. and Yoon, E. (2016). Single cell dual adherent-suspension co-culture micro-environment for studying tumor–stromal interactions with functionally selected cancer stem-like cells. Lab Chip, 16, 2935–2945, doi:10.1039/C6LC00062B.

Chi, L., Zou, Y., Qin, L., Ma, W., Hao, Y., Tang, Y., Luo, R. and Wu, Z. (2017). TIMELESS contributes to the progression of breast cancer through activation of MYC. Breast Cancer Res., 19, 53, doi:10.1186/s13058-017-0838-1.

References 27

Choi, Y., Hyun, E., Seo, J., Blundell, C., Kim, H. C., Lee, E., Lee, S. H., Moon, A., Moon, W. K. and Huh D. (2015). A microengineered pathophysiological model of early-stage breast cancer. Lab Chip, 15, 3350–3357, doi:10.1039/c5lc00514k.

Chung, B., Esmaeili, A. A., Gopalakrishna-Pillai, S., Murad, J. P., Andersen, E. S., Kumar Reddy, N., Srinivasan, G., Armstrong, B., Chu, C., Kim, Y., Tong, T., Waisman, J., Yim, J. H., Badie, B. and Lee, P. P. (2017). Human brain metastatic stroma attracts breast cancer cells via chemokines CXCL16 and CXCL12. npj Breast Cancer, 3, 6, doi:10.1038/s41523-017-0008-8.

Clay, N. E., Shin, K., Ozcelikkale, A., Lee, M. K., Rich, M. H., Kim, D. H., Han, B. and King, H. (2016). Modulation of Matrix Softness and Interstitial Flow for 3D Cell Culture Using a Cell-Microenvironment-on-a-Chip System. ACS Biomater. Sci. Eng., 2, 1968–1975, doi:10.1021/acsbiomaterials.6b00379.

DA Cruz Paula, A., Marques, O., Sampaio, R., Rosa, A., Garcia, J., Rêma, A., DE Fátima Faria, M., Silva, P., Vizcaíno, R. and Lopes, C. (2016). Characterization of CD44+ALDH1+Ki-67- Cells in Non-malignant and Neoplastic Lesions of the Breast. Anticancer Res., 36, 4629–38, doi:10.21873/anticanres.11013.

Daubriac, J., Han, S., Grahovac, J., Smith, E., Hosein, A., Buchanan, M., Basik, M. and Boucher, Y. (2018). The crosstalk between breast carcinoma-associated fibroblasts and cancer cells promotes RhoA-dependent invasion via IGF-1 and PAI-1. Oncotarget, 9, 10375–10387, doi:10.18632/oncotarget.23735.

Drews, J. (2000). Drug discovery: a historical perspective. Science, 287(5460), 1960-4, doi:10.1126/science.287.5460.1960.

Ehteshami Bejnordi, B., Mullooly, M., Pfeiffer, R. M., Fan, S., Vacek, P. M., Weaver, D. L., Herschorn, S., Brinton, L. A., van Ginneken, B., Karssemeijer, N., Beck, A. H., Gierach, G. L., van der Laak, J. A. W. M. and Sherman, M. E. (2018). Using deep convolutional neural networks to identify and classify tumor-associated stroma in diagnostic breast biopsies. Mod. Pathol., 31, 1502–1512, doi:10.1038/s41379-018-0073-z.

Eletxigerra, U., Martinez-Perdiguero, J., Barderas, R., Pingarrón, J. M., Campuzano, S. and Merino, S. (2016). Surface plasmon resonance immunosensor for ErbB2 breast cancer biomarker determination in human serum and raw cancer cell lysates. Anal. Chim. Acta, 905, 156–162, doi:10.1016/j.aca.2015.12.020.

Eslami Amirabadi, H., SahebAli, S., Frimat, J. P., Luttge, R. and den Toonder, J. M. J. (2017). A novel method to understand tumor cell invasion: integrating extracellular matrix mimicking layers in microfluidic chips by “selective curing.” Biomed. Microdevices, 19, 92, doi:10.1007/s10544-017-0234-8.

Farnie, G., Johnson, R. L., Williams, K. E., Clarke, R. B. and Bundred, N. J. (2014). Lapatinib inhibits stem/progenitor proliferation in preclinical in vitro models of ductal carcinoma in situ (DCIS). Cell Cycle, 13, 418–425. doi:10.4161/cc.27201.

Feng, Z.-M., Qiu, J., Chen, X.-W., Liao, R.-X., Liao, X.-Y., Zhang, L.-P., Chen, X., Li, Y., Chen, Z.-T. and Sun, J.-G. (2015). Essential role of miR-200c in regulating self-renewal of breast cancer stem cells and their counterparts of mammary epithelium. BMC Cancer, 15, 645, doi:10.1186/s12885-015-1655-5.

Goel, S., DeCristo, M. J., Watt, A. C., BrinJones, H., Sceneay, J., Li, B. B., Khan, N., Ubellacker, J. M., Xie, S., Metzger-Filho, O., Hoog, J., Ellis, M. J., Ma, C. X., Ramm, S., Krop, I. E., Winer, E. P., Roberts, T. M., Kim, H. J., McAllister, S. S. and Zhao, J. J. (2017). CDK4/6 inhibition triggers anti-tumour immunity. Nature, 548, 471–475, doi:10.1038/nature23465.

Greenbaum, A., Zhang, Y., Feizi, A., Chung, P.-L., Luo, W., Kandukuri, S. R. and Ozcan, A. (2014). Wide-field computational imaging of pathology slides using lens-free on-chip microscopy. Sci. Transl. Med., 6, 267ra175, doi:10.1126/scitranslmed.3009850.

Hainsworth, J. D., Murphy, P. B., Alemar, J. R., Daniel, B. R., Young, R. R. and Yardley, D. A. (2016). Use of a multiplexed immunoassay (PRO Onc assay) to detect HER2 abnormalities in circulating tumor cells of women with HER2-negative metastatic breast cancer: lack of response to HER2-targeted therapy. Breast Cancer Res. Treat., 160, 41–49, doi:10.1007/s10549-016-3969-7.

Advanced Non-animal Models in Biomedical Research: Breast Cancer28

Härmä, V., Schukov, H. P., Happonen, A., Ahonen, I., Virtanen, J., Siitari, H., Åkerfelt, M., Lötjönen, J. and Nees, M. (2014). Quantification of dynamic morphological drug responses in 3D organotypic cell cultures by automated image analysis. PLoS One, 9, doi:10.1371/journal.pone.0096426.

Hohensee, I., Chuang, H.-N., Grottke, A., Werner, S., Schulte, A., Horn, S., Lamszus, K., Bartkowiak, K., Witzel, I., Westphal, M., Matschke, J., Glatzel, M., Jücker, M., Pukrop, T., Pantel, K. and Wikman, H. (2017). PTEN mediates the cross talk between breast and glial cells in brain metastases leading to rapid disease progression. Oncotarget, 8, 6155–6168, doi:10.18632/oncotarget.14047.

Holen, I., Speirs, V., Morrissey, B. and Blyth, K. (2017). In vivo models in breast cancer research: progress, challenges and future directions. Dis. Model. Mech., 10, 359–371, doi:10.1242/dmm.028274.

Ishii, K., Morii, N. and Yamashiro, H. (2019). Pertuzumab in the treatment of HER2-positive breast cancer: an evidence-based review of its safety, efficacy, and place in therapy. Core Evid., 14, 51–70, doi:10.2147/CE.S217848.

Jin, X., Jin, X. and Kim, H. (2017). Cancer stem cells and differentiation therapy. Tumor Biol., 39, 101042831772993, doi:10.1177/1010428317729933.

Ju, J., Zhu, A.-J. and Yuan, P. (2018). Progress in targeted therapy for breast cancer. Chronic Dis. Transl. Med., 4, 164–175, doi:10.1016/j.cdtm.2018.04.002.

Karapanagiotis, S., Pharoah, P. D. P., Jackson, C. H. and Newcombe, P. J. (2018). Development and External Validation of Prediction Models for 10-Year Survival of Invasive Breast Cancer. Comparison with PREDICT and CancerMath. Clin. Cancer Res., 24, 2110–2115, doi:10.1158/1078-0432.CCR-17-3542.

Kinders, R., Parchment, R. E., Ji, J., Kummar, S., Murgo, A. J., Gutierrez, M., Collins, J., Rubinstein, L., Pickeral, O., Steinberg, S. M., Yang, S., Hollingshead, M., Chen, A., Helman, L., Wiltrout, R., Simpson, M., Tomaszewski, J. E., Doroshow, J. H. (2007). Phase 0 clinical trials in cancer drug development: from FDA guidance to clinical practice. Mol Interv., 7(6), 325-34, doi:10.1124/mi.7.6.9.

Kirkizlar, E., Zimmermann, B., Constantin, T., Swenerton, R., Hoang, B., Wayham, N., Babiarz, J. E., Demko, Z., Pelham, R. J., Kareht, S., Simon, A. L., Jinnett, K. N., Rabinowitz, M., Sigurjonsson, S. and Hill, M. (2015). Detection of Clonal and Subclonal Copy-Number Variants in Cell-Free DNA from Patients with Breast Cancer Using a Massively Multiplexed PCR Methodology. Transl. Oncol., 8, 407–416, doi:10.1016/j.tranon.2015.08.004.

Kola, I. and Landis, J. (2004). Can the pharmaceutical industry reduce attrition rates? Nat Rev Drug Discov., 3(8), 711-5, doi:10.1038/nrd1470.

Li, X., Xu, J., Tang, X., Liu, Y., Yu, X., Wang, Z. and Liu, W. (2016). Anthocyanins inhibit trastuzumab-resistant breast cancer in vitro and in vivo. Mol. Med. Rep., 13, 4007–4013, doi:10.3892/mmr.2016.4990.

Liu, K., Newbury, P. A., Glicksberg, B. S., Zeng, W. Z. D., Paithankar, S., Andrechek, E. R. and Chen, B. (2019). Evaluating cell lines as models for metastatic breast cancer through integrative analysis of genomic data. Nat. Commun., 10, 2138, doi:10.1038/s41467-019-10148-6.

Lucantoni, F., Lindner, A. U., O’Donovan, N., Düssmann, H. and Prehn, J. H. M. (2018). Systems modeling accurately predicts responses to genotoxic agents and their synergism with BCL-2 inhibitors in triple negative breast cancer cells. Cell Death Dis., 9, 42, doi:10.1038/s41419-017-0039-y.

Manning, H. C., Buck, J. R. and Cook, R. S. (2016). Mouse Models of Breast Cancer: Platforms for Discovering Precision Imaging Diagnostics and Future Cancer Medicine. J. Nucl. Med., 57, 60S-68S, doi:10.2967/jnumed.115.157917.

McCabe, N., Turner, N. C., Lord, C. J., Kluzek, K., Białkowska, A., Swift, S., Giavara, S., O’Connor, M. J., Tutt, A. N., Zdzienicka, M. Z., Smith, G. C. and Ashworth, A. (2006). Deficiency in the Repair of DNA Damage by Homologous Recombination and Sensitivity to Poly(ADP-Ribose) Polymerase Inhibition. Cancer Res., 66, 8109–8115, doi:10.1158/0008-5472.CAN-06-0140.

References 29

Mylona, E., Melissaris, S., Giannopoulou, I., Theohari, I., Papadimitriou, C., Keramopoulos, A. and Nakopoulou, L. (2014). Y-box-binding protein 1 (YB1) in breast carcinomas: Relation to aggressive tumor phenotype and identification of patients at high risk for relapse. Eur. J. Surg. Oncol., 40, 289–296, doi:10.1016/j.ejso.2013.09.008.

Nagaraju, S., Truong, D., Mouneimne, G. and Nikkhah, M. (2018). Microfluidic Tumor–Vascular Model to Study Breast Cancer Cell Invasion and Intravasation. Adv. Healthc. Mater., 7, doi:10.1002/adhm.201701257.

Park, M. K., Lee, C. H. and Lee, H. (2018). Mouse models of breast cancer in preclinical research. Lab. Anim. Res., 34, 160, doi:10.5625/lar.2018.34.4.160.

Patidar, K., Deshmukh, A., Bandaru, S., Lakkaraju, C., Girdhar, A., Gutlapalli, V., Banerjee, T., Nayarisseri, A. and Singh, S. K. (2016). Virtual Screening Approaches in Identification of Bioactive Compounds Akin to Delphinidin as Potential HER2 Inhibitors for the Treatment of Breast Cancer. Asian Pacific J. Cancer Prev., 17, 2291–2295, doi:10.7314/APJCP.2016.17.4.2291.

Reiter, J. G., Makohon-Moore, A. P., Gerold, J. M., Heyde, A., Attiyeh, M. A., Kohutek, Z. A., Tokheim, C. J., Brown, A., DeBlasio, R. M., Niyazov, J., Zucker, A., Karchin, R., Kinzler, K. W., Iacobuzio-Donahue, C. A., Vogelstein, B. and Nowak, M.A. (2018). Minimal functional driver gene heterogeneity among untreated metastases. Science, 361, 1033–1037, doi:10.1126/science.aat7171.

Ren, X., Ghassemi, P., Kanaan, Y. M., Naab, T., Copeland, R. L., Dewitty, R. L., Kim, I., Strobl, J. S. and Agah, M. (2018). Kernel-Based Microfluidic Constriction Assay for Tumor Sample Identification. ACS Sensors, 3, 1510–1521, doi:10.1021/acssensors.8b00301.

Riching, K. M., Cox, B. L., Salick, M. R., Pehlke, C., Riching, A. S., Ponik, S. M., Bass, B. R., Crone, W. C., Jiang, Y., Weaver, A. M., Eliceiri, K. W. and Keely, P. J. (2014). 3D collagen alignment limits protrusions to enhance breast cancer cell persistence. Biophys. J., 107, 2546–2558, doi:10.1016/j.bpj.2014.10.035.

Richmond, A. and Su, Y. (2008). Mouse xenograft models vs GEM models for human cancer therapeutics. Dis. Model. Mech., 1, 78–82, doi:10.1242/dmm.000976.

Ross, J. S., Slodkowska, E. A., Symmans, W. F., Pusztai, L., Ravdin, P. M. and Hortobagyi, G. N. (2009). The HER-2 Receptor and Breast Cancer: Ten Years of Targeted Anti-HER-2 Therapy and Personalized Medicine. Oncologist, 14, 320–368, doi:10.1634/theoncologist.2008-0230.

Sabhachandani, P., Motwani, V., Cohen, N., Sarkar, S., Torchilin, V. and Konry, T. (2016). Generation and functional assessment of 3D multicellular spheroids in droplet based microfluidics platform. Lab Chip, 16, 497–505, doi:10.1039/c5lc01139f.

Shirure, V. S., Bi, Y., Curtis, M. B., Lezia, A., Goedegebuure, M. M., Goedegebuure, S. P., Aft, R., Fields, R. C. and George, S. C. (2018). Tumor-on-a-chip platform to investigate progression and drug sensitivity in cell lines and patient-derived organoids. Lab Chip, 18, 3687–3702, doi:10.1039/c8lc00596f.

Smith, M. T., Guyton, K. Z., Gibbons, C. F., Fritz, J. M., Portier, C. J., Rusyn, I., DeMarini, D. M., Caldwell, J. C., Kavlock, R. J., Lambert, P. F., Hecht, S. S., Bucher, J. R., Stewart, B. W., Baan, R. A., Cogliano, V. J. and Straif, K. (2016). Key Characteristics of Carcinogens as a Basis for Organizing Data on Mechanisms of Carcinogenesis. Environ. Health Perspect., 124, 713–721, doi:10.1289/ehp.1509912.

Sung, W., Jeong, Y., Kim, H., Jeong, H., Grassberger, C., Jung, S., Ahn, G. O., Kim, I. H., Schuemann, J., Lee, K. and Ye, S. J. (2018). Computational Modeling and Clonogenic Assay for Radioenhancement of Gold Nanoparticles Using 3D Live Cell Images. Radiat. Res., 190, 558, doi:10.1667/RR15134.1.

Vici, P., Mottolese, M., Pizzuti, L., Barba, M., Sperati, F., Terrenato, I., Di Benedetto, A., Natoli, C., Gamucci, T., Angelucci, D., Ramieri, M. T., Di Lauro, L., Sergi, D., Bartucci, M., Dattilo, R., Pagliuca, A., De Maria, R. and Maugeri-Saccà, M. (2014). The Hippo transducer TAZ as a biomarker of pathological complete response in HER2-positive breast cancer patients treated with trastuzumab-based neoadjuvant therapy. Oncotarget, 5, 9619–9625, doi:10.18632/oncotarget.2449.

Advanced Non-animal Models in Biomedical Research: Breast Cancer30

WHO (2020). Incidence of female breast cancer per 100 000. Eur. Heal. Inf. Gatew. Available at: https://gateway.euro.who.int/en/indicators/hfa_375-2350-incidence-of-female-breast-cancer-per-100-000/ [Accessed June 3, 2020].

Xia, J.-L., Fan, W.-J., Zheng, F.-M., Zhang, W.-W., Xie, J.-J., Yang, M.-Y., Kamran, M., Wang, P., Teng, H.-M., Wang, C.-L. and Liu, Q. (2017). Inhibition of AURKA kinase activity suppresses collective invasion in a microfluidic cell culture platform. Sci. Rep., 7, 2973, doi:10.1038/s41598-017-02623-1.

Yang, Y., Yang, X., Zou, J., Jia, C., Hu, Y., Du, H. and Wang, H. (2015). Evaluation of photodynamic therapy efficiency using an in vitro three-dimensional microfluidic breast cancer tissue model. Lab Chip, 15, 735–744, doi:10.1039/c4lc01065e.

Yardley, D. A., Noguchi, S., Pritchard, K. I., Burris, H. A., Baselga, J., Gnant, M., Hortobagyi, G. N., Campone, M., Pistilli, B., Piccart, M., Melichar, B., Petrakova, K., Arena, F. P., Erdkamp, F., Harb, W. A., Feng, W., Cahana, A., Taran, T., Lebwohl, D. and Rugo, H. S. (2013). Everolimus Plus Exemestane in Postmenopausal Patients with HR+ Breast Cancer: BOLERO-2 Final Progression-Free Survival Analysis. Adv. Ther., 30, 870–884, doi:10.1007/s12325-013-0060-1.

Yildiz-Ozturk, E., Gulce-Iz, S., Anil, M. and Yesil-Celiktas, O. (2017). Cytotoxic responses of carnosic acid and doxorubicin on breast cancer cells in butterfly-shaped microchips in comparison to 2D and 3D culture. Cytotechnology, 69, 337–347, doi:10.1007/s10616-016-0062-3.

Zhong, Z., Rosenow, M., Xiao, N. and Spetzler, D. (2018). Profiling plasma extracellular vesicle by pluronic block-copolymer based enrichment method unveils features associated with breast cancer aggression, metastasis and invasion. J. Extracell. Vesicles, 7, 1458574, doi:10.1080/20013078.2018.1458574.

Zimmer, A. S. and Denduluri, N. (2019). When to Add Additional Anti-HER2 Therapy to Adjuvant Trastuzumab. Curr. Oncol. Rep., 21, 109, doi:10.1007/s11912-019-0848-5.

Advanced Non-animal Models in Biomedical Research: Breast Cancer32

Table 1: Inclusion criteria used to retrieve scientific articles from literature.

1. Cells cultures and/or co-cultures in 2D, 2.5D, 3D or Microphysiological Systems (MPS)

a. Primary cell cultures

b. Immortalised cell lines

c. Stem cells (SCs)

i. Pluripotent SCs

• Induced pluripotent SCs (iPSCs)

• Embryonic SCs (ESCs)

ii. Multipotent SCs

• Somatic SCs

• Fetal SCs

2. Ex vivo material

a. Biopsies

b. Organotypic cultures

c. Stem cells (SCs)

i. Explants

ii. Whole organ or organ slice

3. Cell-free assays

a. Biochemical assays

4. Gene reporting assays

5. In silico

a. Algorithm

b. Mathematical

c. Computational

d. Simulations

6 Annex

Annex 33

1. The study does not deal with breast cancer

2. Secondary literature (review, meeting abstract, etc.)

3. Duplicate

4. No in vitro or in silico model or method

5. In vivo study

6. Test method not able to measure endpoints

7. The study does not focus on development/characterization of a valuable alternative test method/model

8. No information on applications

9. The study does not provide mechanistic/pathophysiological or biological relevance

10. No biomedical research application

11. No valuable non-animal model or method

12. Not English articles

13. Retracted Publication

14. Published before 2014

Table 2: Exclusion criteria used to retrieve scientific articles from literature.

Advanced Non-animal Models in Biomedical Research: Breast Cancer34

Multidisciplinary citation databases and indexing services

Web of Science http://webofknowledge.com/WOS

Scopus http://www.scopus.com/

Google Scholar http://scholar.google.com/

Biomedical sciences citation databases

PubMed http://www.ncbi.nlm.nih.gov/pubmed

Foundations

American Breast Cancer Foundation http://www.abcf.org/

Breast Cancer Research Foundation http://www.bcrf.org/

Breastcancer.org http://www.breastcancer.org/

Societies

Breast Cancer – American Cancer Society https://cancer.org/cancer/breast-cancer.html

Eusoma http://www.Eusoma.org

Research institutions and programs

NIH – National Cancer Institute https://www.cancer.gov/types/breast/research

California Breast Cancer Research Program http://cbcrp.org

Three-dimensional breast cancer models for X-ray Imaging research – MaXIMA

https://maxima-tuv.eu/

Breast Cancer Research Program | Vanderbilt-Ingram Cancer Center

https://www.vicc.org/research/programs/breast

Westmead Breast Cancer Institute https://www.bci.org.au

Fred Hutch https://www.fredhutch.org/en/diseases/breast-cancer.html

Charities

Breast Cancer Care https://www.breastcancercare.org.uk/

Breast Cancer Now – the UK’s largest breast cancer research charity

https://breastcancernow.org

Pink Ribbon Foundations www.pinkribbonfoundation.org.uk

Events

European Breast Cancer Conference (EBCC) www.ecco-org.eu/EBCC

Advanced Breast Cancer - 5th ESO-ESMO International Consensus Conference

http://www.abc-lisbon.org

IABCR 2019 – 31st International Association for Breast Cancer Research Conference

https://www.eacr.org/meeting/iabcr-2019-31st-international-association-for-breast-cancer-research-conference

4th Breast Cancer Congress www.breastanbul.org

10th Euro Breast Cancer Summit https://eurobreastcancer.cancersummit.org/2019

Melbourne International Breast Congress https://melbournebreast2018.org

Table 3: Information sources used for literature searches.

Annex 35Advanced Non-animal Models in Biomedical Research: Respiratory Tract Diseases

Strategy Definition Presence rate1 Inclusion rate2 Pros & cons

A Bottom-Up

Wide-range strategy without any search term exclusion. This strategy retrieves a large amount of publications.

It should guarantee the highest presence rate among the bottom-up strategies.

Inclusion rate could be low.

A large amount of publications must be screened.

BBottom-Up + Scoring system

Wide-range strategy followed by a ranking system based on search terms scores.

Top rank should have the higher presence rate than the intermediate and low ranks.

This strategy should concentrate the eligible publications into the top rank (score >200).

Higher amount of publications than Strategy A; however the absolute number of publications could be lower.

CBottom-Up + Top-down + Scoring system

Wide-range strategy followed by a ranking system based on search terms scores. The exclusion terms are tailored by analysing eligible publications specific for each lot.

This strategy concentrated the eligible publications into a top-ranking class (score >200).

Top rank should have higher inclusion rate than Strategy C.

Higher publications than in Strategy A.

D Stand-by

The most represented redundant models are actively searched and shelved.

Not applicable Not applicable

It avoids information dilution of new models, which are underrepresented. On the contrary, some applications of most represented models can be lost.

E EnrichmentSpecific search terms for new models’ retrieval.

Not applicable Not applicable

It enriches the search with underrepresented models.

1 Presence rate: Percentage of pre-selected eligible publications existing inbuilt dataset for each lot.

2 Inclusion rate: Percentage of eligible publications selected by title and abstract analysis.

Table 4: Organisations relevant to respiratory disease modelling and non-animal methods.

Advanced Non-animal Models in Biomedical Research: Breast Cancer36

Field Definition Drop-down option

Model numberModel of breast cancer which is described in a paper

NA

Breast cancer type Type of breast cancerBC (Breast Cancer) IBC (Inflammatory Breast Cancer) BC/IBC

Breast cancer subtypeMorphological, molecular and clinical characteristics investigated in the model

For example: BC (breast cancer) Claudin-low BC DCIS (Ductal Carcinoma in Situ) IBC (Inflammatory Breast Cancer) IDC (Invasive Ductal Carcinoma) IDC/ILC ((Invasive Ductal Carcinoma/Invasive Lobular Carcinoma)

Disease features The disease feature studied by the model

For example: Cancer initiation and development Angiogenesis Metastasis Cancer landscape Drug discovery and/or testing

CategoryThe category of non-animal model assigned to the model

In vitro In silico In vitro/in silico

Type More specifications of the model category

Cells Cell-free Ex vivo Computational Algorithm Simulation Mathematical and their combinations

Cells Biological material source, if any

Immortalised Primary Stem cells n/a

Source Biological material source, if any

MCF-7 MDA-MB-231 fibroblasts and endothelial cells liquid biopsy

Cell culture typeIf the model employs cells, this field specifies the tpe of cell culture

Culture Co-culture MPS (Microphysiological systems) and their combinations

Cel culture dimensionsIf the model employs cells, this field specifies the dimensions of the cell culture

2D 2.5D 3D and their combinations

3DIf the model uses 3D cell cultures, this field specifies the type of the 3D dimension

Scaffolds Spheroids Organoid and their combinations

Table 5: Agreed categories for data extraction.

Annex 37

Ex vivoIf the model is based on ex vivo cells/tissues, this field specifies the type of materials

Biopsies Liquid biopsies Organ slice Whole organ and their combinations

ApplicationsMain scientific aim or application of the model

For example: Diagnosis of diseases Model/method development Diseases mechanism Drug development/testing

Biological endpointsList of potential biological endpoints used in a model system to describe the disease mechanism and/or study focus

For example: Cytotoxicity Cell proliferation Invasion Metabolic activity

ThroughputRegarding productivity/automatisation of the model

High Medium Low

PotentialPossible multiple model application in addressing disease features

Yes (The method/model has future potential for its breast cancer applications). No (The method/model has no future potential for its breast cancer applications). n/a (not specified)

Relevance Biological relevance of the model for the disease feature in replacing animal models

Direct (The model is sufficient for the conclusions of the study). Supportive (The model is partially supporting the conclusions of the study). n/a (not specified)

DOI or linkDigital Object Identification number to retrieve the publication abstract. If not available, an alternative link is provided

-

First author nameName of the first author of the peer-reviewed article

-

Year Publication year from 2014 to 2019 -

GETTING IN TOUCH WITH THE EU

IN PERSONAll over the European Union there are hundreds of Europe Direct information centres. You can findthe address of the centre nearest you at: https://europa.eu/european-union/contact_en.

ON THE PHONE OR BY EMAILEurope Direct is a service that answers your questions about the European Union. You can contactthis service:- by freephone: 00 800 6 7 8 9 10 11 (certain operators may charge for these calls),- at the following standard number: +32 22999696, or- by electronic mail via: https://europa.eu/european-union/contact_en.

FINDING INFORMATION ABOUT THE EU

ONLINEInformation about the European Union in all the official languages of the EU is availableon the Europa website at: https://europa.eu/european-union/index_en.

EU PUBLICATIONSYou can download or order free and priced EU publications from EU Bookshop at: https://publications.europa.eu/ en/publications. Multiple copies of free publications may be obtained by contacting Europe Direct or your local information centre (see https://europa.eu/european-union/contact_en).

EU LAW AND RELATED DOCUMENTSFor access to legal information from the EU, including all EU law since 1952in all the official language versions, go to EUR-Lex at: http://eur-lex.europa.eu.

OPEN DATA FROM THE EUThe EU Open Data Portal (http://data.europa.eu/euodp/en) provides access to datasets from the EU. Data can be downloaded and reused for free, for both commercial and non-commercial purposes.

JRC MissionAs the science and knowledge service of the European Commission, the Joint Research Centre’s mission is to support EU policies with independent evidence throughout the whole policy cycle.

@EU_ScienceHub

EU Science Hub - Joint Research Centre

EU Science, Research and Innovation

EU Science Hub

EU Science Hub ec.europa.eu/jrc

The European Commission’s science and knowledge service Joint Research Centre

KJ-NC-30334-EN

-N

doi:10.2760/618741ISBN 978-92-76-24689-3